Design Experiments Details

Study 1: Proof of concept

• For the first trial of the concept, I decided to put two big eyes on the building facade which would follow people as they walked across
• First was the Low resolution LED Design, then the High resolution Projection Design

• I found that the location of the installation being on the side of a passageway and up high could have made it hard to grab attention
• People could not tell what the LED design was showing, but the High Resolution Projection eyes were easily recognised and attracted a bit more attention
• Most said they thought it was a bit creepy to have eyes following them
• Since the lights were on every night (with a different design), people didn't realise that these eyes were not normally there

Baseline Study

• We spent a night watching the behaviours when the usual content was displayed (since this facade normally has abstract low-res patterns then video projections everynight)
• The aim was to obtain interaction counts to have a baseline to compare with

• There was a slightly higher interaction count compared to study one
• From the observations I had two hypotheses: that the more colourful designs attracted attention and that the changing nature of the designs retained more attention - these two aspects differed from the design in study 1

Study 2: Adding Curiosity Objects + Face

• Since we found that people didn't suspect anything different about the design compared to the usual lights, I decided to add "Curiosity Objects"
• Curiosity objects as described in literature, are aimed to attract curiosity and attention - I made paper ghost cutouts and put a cloth over a statue with two paper eyes
• I also added a Face/Body to the Ghost eyes to make it less creepy as the intention was to make it fun

• The curiosity objects, particularly the Ghost cloth statue attracted a lot of attention
• People noted that it caught their attention because "It's not usual"
• Perceptions of the ghost were "Cute" and "Cartoon like" rather than "Creepy

Study 3: Adding Sound with the Curiosity Objects

• For the next experiment I wanted to see if adding sounds and music would affect interpretations or impact the interaction counts
• So along with the curiosity objects, speakers played music from a "fantasy magical spooky music" spotify playlist and "ping" sounds when the eyes blinked

• Sounds greatly increased the ‘glance’ count
• The sounds and music seemed to have influenced interpretation e.g. mentions of “haunted”, “spooky”, “party"
• Low-res design was still difficult to interpret

Hybrid Resolution Design Iterations

I conducted 3 iterations through 5 Studies. For each study, I split the hour into 30 minutes each for the 2 concepts. Only Concept 1 (Light Shadow) used an X-box Kinect to detect motion, the second concept was kept as manually controlled but the Kinect was left in front of the installation.

Defining the Goal Experience

Although the main aim is to conduct research, I also focused on making it into a great experience. Thus from part one, I defined 2 types of passers-by to cater for:

Regular

• Regularly walks past the area
• Unamused by current installation as it is always there but does like it as "colourful lights"

First time

• Walks past the area for the first time
• Unfamiliar with the university and notices the installation, usually wondering if it is a permanent installation or temporary

I then also defined keywords to describe the experience I hope to create for them:

• Fun - having no particular purpose other than bringing some joy
• Engaging - people would interact with it, more than a glance
• Surprise - to be recognised as unusual and out of people's expectations of the installation
• Curiosity - create a sense of wonder and prompt people to question it

Creating the Experience

There were also study "pairs" where I used the same designs across two studies but the difference was whether I promoted the study as an event or not (through event posters around campus and social media) - to answer the third research question - Will behaviours and perception differ when people come to the area knowing about the Installation?

Similar to Part 1, I conducted onsite semi-structured interviews and interaction counts were recorded. Findings were analysed and changes implemented for each iteration as seen below.

Summary of Design considerations and findings

• I decided to keep the curiosity objects (ghost Stickers, ghost statue and the audio) from part 1 for most of the sessions as people noted that it added to the vibe and particularly the music was a large influence to attract attention - I decided on the "spooky magical" playlist over a more "haunted" playlist as the ghost was not a scary design
• For concept 1 (Light shadow), having the light shadow position more accurately align to the position of the user (through technical iterations) greatly impacted understanding of it
• For concept 2 (Hybrid ghost) the initial change from only the ghost eye pupils following passersby to having the whole ghost follow the passersby made a big difference to the understanding of the Low-resolution "body" being connected to the High resolution "head"
• I then added colour to the design; red, blue and Green themes that changed at random intervals which retained attention. As one interviewee who stopped for a while explained “I was looking at the colours and I was like how many, is it just looping like the one thing?” and after being explained the concept they said “I want to keep watching it, because it’s not on a loop"
• I also made the pupils change to a heart shape if people interacted with it - to further enhance the idea that some people said it could be like a companion - the change saw more people ask questions about it as if it was alive like “does it come out every night?” rather than previously (no heart pupil) “is this here every night?”. One stated that it showed “friendliness towards the people…if I come here to chill, I will feel like as if I have a companion nearby so I won't be lonely”.
• The "ping" sounds that triggered when the eyes blinked were changed to a cute "pop" sound as people expressed previously that the"game" sound didn't fit with the friendly ghost theme.
• As the studies drew closer to Halloween day, people started to guess that it was for halloween and recognise it as a ghost

Hybrid Resolution Design Iteration Details

Study 4: Hybrid Resolution Version 1 (Unpromoted)

For both concepts, the Ghost stayed at the middle of the wall with small "floating" movements while the pupils would follow the passerby as they walked across.

• There were technical difficulties on the night such as the music failing to play and the kinect tracking was not working as intended so it was switched to manual.
• Although the music failed to play, the sound "ping" still played when the ghost blinked. This saw the "ping" sound to be more prominent without the music where the interact rate doubled as people thought "there was a party or game somewhere"
• The main problem I discovered was that people did not perceive the high and low-resolution parts to be working together, but rather as separate parts where LED was permanent whereas the projection was the newer part that followed people (eyes) - so for the next iteration I wanted to make the whole ghost move to emphasise the continuous flow as the lower part would have movement too

‍

Study 5 (Unpromoted) & 6 (Promoted event): Hybrid Resolution Version 2

For Light Shadows, the high-res head moved left or right left or right as detected by the xbox kinect while the low-res showed a "shadow" of passer-bys as they walked across.

For the hybrid ghost the whole ghost (high-res "head" with low-res "body") moved as controlled manually to follow passer-bys.

• For the hybrid ghost, people were able to perceived the two resolutions as related. Particularly when the ghost moved down, going out of view of the projection where the "eyes" moved into the-low area, people were able to see it.
• Surprisingly, for the light shadows, more people realised that they were being followed, but when asked about the light shadow, the majority said they did not notice the LEDs suggesting an unconscious effect that it may have had

‍

Study 7 (Unpromoted) & 8 (Halloween Event): Hybrid Resolution Version 3

For both concepts, I added an animated stars background that changed colours at random intervals and the pupils would turn heart shaped as controlled by us whenever someone interacts with it. The original "ping" sounds for blinking was changed to a cure "pop" sound that match the heart pupil change.

• The colour and animation were noted by interviews as key-points of what made it appealing to them
• Particularly those who had seen the ghost from my previous studies, they found it more "interesting" and "nice"
• The changing colours also meant many wondered "how many colours" or "what other shapes" there are as they stopped to look at it, confirming my hypothesis that the changing nature can retain attention
• The heart pupils also led to more interviewees describing the ghost to have "personality" and emotion saying "it loves me" - the pop sound was also said to add to the "cuteness" of the ghost
• The concept of the "Emerging champion" as described was also observed, where previous interviewees who frequented the are would point out the ghost and its story to their friends
• These last two studies were also close to Halloween, most were able to guess that it was for Halloween which added to the vibe and recognition of it as a"Ghost" for "Halloween"

‍